Normally, cell proliferation is regulated by both growth-promoting signals and growth-constraining signals. These two kinds of signals for each cell would normally strike a balance in a manner which reflects the need of the body for the particular cell. If a cell fails to respond to the growth-constraining signals or over-responds to the growth-promoting signals, it will proliferate abnormally fast (referred to as neoplastic cells) and may eventually develop into cancer, a malignant neoplasm.
Chemotherapy, a current method of treating cancer, is generally based on the fast-proliferating property of cancer cells. Since cancer cells proliferate rapidly, they are more sensitive to drugs which inhibit cellular proliferation. In theory, by carefully choosing the dosage of chemotherapeutic drugs, one can inhibit cancer cell proliferation without seriously damaging normal cells. However, some normal cells, such as hematopoietic stem cells, also proliferate rapidly. Therefore, any dosage which is harmful to cancer cells is often also harmful to the hematopoietic stem cells. On the other hand, if the dosage is not high enough to kill the cancer cells, there is the risk that the cancer would reappear shortly after chemotherapy is terminated.
Because it is hard to find a dosage which selectively kills cancer cells, high-dose chemotherapy followed by autologous hematopoietic progenitor stem cell transplantation has gained extensive application as a therapeutic approach in many cancers (see, for example, Winter, 1999; Nieto and Shpall, 1999). In this approach, a portion of the hematopoietic stem cells is removed from a cancer patient, and the patient is then treated with high-dose chemotherapy which is lethal to rapid-proliferating cells, such as cancer cells and hematopoietic stem cells. Subsequently, the patient receives transplantation of autologous hematopoietic stem cells which have been previously removed from the same patient to regenerate the hematopoietic system.
A serious drawback of this therapy is that when the hematopoietic progenitor stem cells are removed from the patients, they are often contaminated with cancer cells. This is especially a problem when the patient has a cancer of the hematopoietic origin, but patients with a solid tumor may also suffer from contamination of the hematopoietic stem cells, particularly if the solid tumor has metastasized. As a result, when the removed cells are transplanted back to reestablish the hematopoietic system, some cancer cells may also be placed back to the cancer patient where they may proliferate again and contribute to cancer recurrence. It is therefore desirable to purge the autografts before transplantation.
Several methods have been employed to purge autographs (Spyridonidis et al., 1998; Bensinger 1998). The autograph can be treated with chemotherapy to kill the contaminating neoplastic cells in vitro. However, as discussed above, it is hard to find a dosage for the chemotherapeutic drug which selectively kills neoplastic cells or cancer cells but leaves normal hematopoietic stem cells intact. Autographs can also be treated with a toxin conjugated to antibodies which recognize an antigen that is specific to the neoplastic cells, but such a tumor specific antigen does not always exist.
It is also possible to separate stem cells from the other cells based on a stem cell specific surface marker (CD34) by using flow cytometry, affinity columns or magnetic beads. However, by selecting only certain hematopoietic cells, e.g., the CD34+ cells, other hematopoietic cells such as T cells, B cells, monocytes and natural killer cells are also eliminated, and immune recovery may be delayed (Bensinger, 1998). This method also results in the loss of about half the CD34+ cells and retention of some contaminating cancer cells (Spyridonidis et al., 1998). Therefore, there remains a need for a highly selective method with a reasonable yield to purge autografts which may contain neoplastic cells.